1. Field of the Invention
This invention relates to a thin-film magnetic head.
2. Description of the Related Art
In recent years, the densification of magnetic recording has advanced to a point where systems of such high recording density as 500 Mb/inch.sup.2 in VTR and 200 Mb/inch.sup.2 in HDD have already found acceptance for practical use. Demands for further densification of magnetic recording know no bound. In consequence of the advancing densification of recording, the decrease of track width has become an essential task. The HDD of a recording density of 200 Mb/inch.sup.2, for example, has sufficiently fulfilled its purpose when the width of a track is 7 .mu.m the interval between tracks is about 2 .mu.m, and the tolerance of track width is roughly equal to the interval between tracks (2 .mu.m). In order that the HDD may allow further densification of recording, it becomes necessary to decrease the track width to below 5 or 6 .mu.m and the tolerance to below 0.5 .mu.m. Further, it is suspected that the densification of recording to the order of 10 Gb/inch.sup.2 will require the track width to be not more than 1 .mu.m and the tolerance to be in the neighborhood of 0.1 .mu.m. To satisfy these requirements, the magnetic head calls for a distinct improvement.
FIG. 11 and FIG. 12 illustrate the construction of a conventional thin-film magnetic head. In these diagrams, 1 denotes a substrate which is formed of such as Al.sub.2 O.sub.3.TiC. On this substrate 1, a first magnetic layer which is made as of Permalloy and destined to serve as a lower magnetic core 3 is superposed through the medium of an insulating layer 2 made as of Al.sub.2 O.sub.3. On the lower magnetic core 3 are superposed a magnetic gap 4 made as of SiO.sub.2 and an insulating layer 6 made as of polyimide and having embedded therein a coil 5 made as of Cu. On the insulating layer 6, a second magnetic layer destined to serve as an upper magnetic core 7 is formed. A protective layer 8 made as of Al.sub.2 O.sub.3 is formed on the substrate 1 including the upper side of the upper magnetic core 7.
With respect to the conventional thin-film magnetic head which is provided with such lower magnetic core 3 and such upper magnetic core 7 as are shaped as described above, it has been pointed out that if the width of a track thereof is decreased to the neighborhood of 4 .mu.m, for example, the magnetic head will be at a disadvantage in tending to orient the axis of easy magnetization thereof in the direction perpendicular to the direction of track width, entailing a decline in the permeability thereof, and inducing degradation of the head characteristics, particularly high-frequency response characteristics "Magnetic Properties of a Narrow-Stripe C--Ta--Zr Amorphous Film," written by Hiroshi Mitsuya et al. and published in the Journal of Japan Applied Magnetics Society, 12, 255-258 (1988)!.
The impartation of a track width to the conventional thin-film magnetic head which is constructed as illustrated in FIG. 11 and FIG. 12 is accomplished by selective plating as with Permalloy. In this case, it is the accuracy with which a resist is patterned in the PEP process (photoengravement process) prior to the selective plating that determines the accuracy of the track width. In the case of the thin-film magnetic head, the difference of level h embracing the underlying layers prior to the formation thereon of the upper magnetic core 7 is about 10 .mu.m. In order for this difference of level to be thoroughly covered, the resist requires a thickness equalling this difference of level. For all the possible devices available for coating, the resist requires a thickness of at least about 5 .mu.m.
When the exposure is carried out by the contact method, the distance between the surface of a photomask and the bottom surface of the resist is about 15 .mu.m (10 .mu.m difference of level +5 .mu.m thickness of resist) at least. The size of an unfocussed spot (the range in which the intensity of light changes from 100% to 50%) due to the Fresnel diffraction which is attendant on the contact exposure of this nature is found by calculation to be 3.5 .mu.m, an unfitting magnitude even for the density of 200 Mb/inch.sup.2 interval between tracks (guard band) about 2 .mu.m!, let alone the accuracy of track width to be achieved in the distant future.
As a solution to this problem, the method of effecting the impartation of a track from the side (ABS layer side) facing the medium has been proposed (IEEE Transaction on Magnetics, Vol. 24, No. 6, November 1988, pp. 2841-2843). Similarly, the method of effecting the impartation of a track from the surface facing the medium by the FIB (focussed ion beam) etching technique has been proposed (refer to JP-A-03-296,907). Though these methods are capable of securing the accuracy of track width as aimed at, they entail a serious problem in terms of the adaptability for mass production because the individual heads on a production line must be processed one by one and because the throughput of FIB's themselves is unusually poor.
Further, the method which aims to narrow the tolerance of processing by forming the upper magnetic core 7 in the form of the combination of a front body 7a vested with the general appearance of a fan with a rear body 7b extended as far as the rear gap part and shaping the leading half part by the ion beam milling technique in the early stage of manufacture in which the difference of level across the underlying layers is relatively small has been proposed (IEEE Transaction on Magnetics, Vol. 27, No. 6, November 1991, pp. 4936-4938). So long as the front body 7a of the upper magnetic core 7 is fabricated by the ion beam milling technique, however, the difference between the size of the resist mask and that of the finished magnetic core is inevitably large and is hardly fit for the density of 10 Gb/inch.sup.2. So long as this method relies for the shaping in question on the ion beam milling technique, it cannot contribute to enhance the productivity because the rate of the milling work is as low as some tens of nm/min. Further, when the ion beam milling technique involves the introduction of a fluorine type gas, the method suffers from a marked decline of productivity because various deposits occur inside the ion beam gun and prevent the gun from being stably used for a long time.
The method which resides in utilizing a selective plating technique for the formation of the front body 7a in the general shape of a fan has been also known. So long as the selective plating technique is used, however, the thickness of the resist mask and the difference of level h' do not deserve to be disregarded. Since the resist must be patterned at least in a depth closely approximating the total thickness (about 6 .mu.m) of the lower and the upper magnetic core, the tolerance of the accuracy of patterning of the resist itself is inevitably enlarged. Besides, very minute portions of the front body 7a are given the desired selective plating only with difficulty.
In any of the methods described above, the width of the upper and the lower magnetic core opposed to each other across the gap (the upper and lower pole width) cannot be completely equalized because the alignment error occurring between the upper and the lower magnetic core demands due respect and, therefore, the upper magnetic core must be decreased from the lower magnetic core by a margin roughly twice as large as the alignment error. Even in the absence of misalignment, the problem of possibly aggravating the burden of side righting will ensue.
As described above, the construction of the thin-film magnetic head and the technique for the impartation of a track to the magnetic head which have been known to the art to date are hardly fit for such accuracy of track width as 10 Gb/inch.sup.2 because they have limits of their own in narrowing the interval between tracks. Thus, the art has not yet realized a thin-film magnetic head of a narrow track interval which satisfies both the requirements regarding dimensional tolerance and adaptability for mass production. The desirability of improving the existing thin-film magnetic head, therefore, has been finding enthusiastic recognition.